Machine learning prediction of 30-day all-cause mortality risk factors in HCC rupture - Takeaways - MDSpire

Machine learning prediction of 30-day all-cause mortality risk factors in HCC rupture

  • By

  • Shixiong Shi

  • Canbin Xie

  • Lin Long

  • June 23, 2026

  • 0 min

Share

  • 1

    Ruptured hepatocellular carcinoma (HCC) significantly contributes to early mortality, accounting for 25%-75% of fatalities in affected patients.

  • 2

    A retrospective study of 156 patients with ruptured HCC utilized machine learning to identify predictors of 30-day mortality.

  • 3

    The decision tree model achieved a sensitivity of 87.5% and an AUC of 0.7901 for predicting 30-day mortality in these patients.

  • 4

    Key predictors identified through SHAP analysis included total bilirubin (TBIL) and INR, with nonlinear effects on mortality risk.

  • 5

    The integration of TBIL and INR improved predictive accuracy, achieving an AUC of 0.87 for assessing 30-day mortality risk.

Original Source(s)

Machine learning prediction of 30-day all-cause mortality risk factors in HCC rupture

  • by Shixiong Shi, Canbin Xie, Lin Long

  • June 23, 2026

Related Content

Severe asparaginase-associated hypertriglyceridemia in pediatric acute lymphoblastic leukemia: a single-center experience

Ultrasound-guided, nursing-led, phase-based swallowing rehabilitation in elderly patients with MIBG-positive medullary thyroid carcinoma following 131I-MIBG therapy

Development and validation of an intra-tumoral and peri-tumoral radiomics model based on dynamic contrast-enhanced ultrasound for predicting lymph node metastasis in type 2 diabetic patients with thyroid cancer